Human Resource Management

APPRAISAL: A PROCESS IN SEARCH OF A TECHNIQUE 1

Reappraisal of appraisal has been an enormous topic throughout the public sector in recent years. First, there has been a quest to provide more objectivity in the process. While this chapter can provide a body of knowledge and dozens of concrete tips, it does also require organizational and managerial resources to some degree as well, at a time when public organizations are suffering resources losses. Second, as the examples in this lecture will indicate, the inclusion of results-oriented measures has required a major reassessment. While few would argue for the types of results-oriented systems that were typical of sales people in the past who were paid 100% on commission, the idea that there should only be trait and behavioral measures—as was common in the past—has been questioned everywhere. As we will see, however, those working in the public sector have numerous types of goals to achieve, where success is as tied to well-crafted and properly funded policies as it is with implementation. Thus, process measures as well as results measures cannot generally be ignored.

INTRODUCTION (2)

So, the employee is recruited, selected and hired, classified, trained, and paid. How are they then evaluated or appraised? The importance of appraisal cannot be underestimated. Appraisal assists with many important functions.

First, it provides enormously powerful feedback to employees to aid in their individual improvement. This necessarily links to their goal setting, training, development, experiential learning and might link to coaching as well.

Second, it provides a way to make differential decisions about rewards such as compensation (evaluation).

Third, it provides an opportunity to enforce standards (compliance and discipline).

Fourth, it provides feedback about the functioning of the system, because an employee’s weak performance might have less to do with their lack of ability and effort than a system weakness such as poor training, lack of direction, lack of supervision, insufficient resources, inappropriate or inefficient organization of work, etc. As the cartoon illustrates, of course, it can be as difficult for those in charge to use the indirect feedback that they get, as for those receiving the direct feedback! Yet when a culture of honest appraisal exists, which is the foundation of the learning organizations about which we spoke in the last chapter, and organizations which have a substantial merit based allocation system discussed in the compensation chapter, feedback is given and taken as an opportunity rather than as a critique.

Broadly speaking, then, organizations that do not provide appraisal in terms of robust feedback loops cannot be top-performing, and may struggle to be mediocre. Nonetheless, the challenges in building good feedback systems are daunting.

3 So what are the paradoxes of appraisal? Fundamentally, the four positive functions mentioned—development, evaluation, standards compliance, and systems improvement—have different purposes and affect employees’ motivation and comfort in radically divergent ways. Ideally, there would be different processes, one for each function, and some organizations do just this, but that consumes more resources and requires more systems discipline to implement and maintain. So to be more efficient, the same system may be used for a variety of purposes such as development, evaluation, and standards compliance. However, such a system will never work ideally for any specific function. Sometimes the tension among the functions built into multipurpose systems will be easily accommodated by managers; however, sometimes such tensions seem nearly impossible to resolve. For example, creating a friendly, supportive open-ended developmental atmosphere, and at the same time making it clear that the end result of the process is a decision that will affect one’s raise or lack thereof can seem nearly irresolvable. Our cartoon illustrates how this may feel to employees.

A second major challenge when looking at the critical functions of evaluation and standards compliance are the psychological, political, and systemic distortions that are endemic when making complex assessments about what other humans do. Consider, for instance, the challenge of providing a succinct review of someone’s complex job. The average job includes about 2000 hours, and thus hundreds if not thousands of completed tasks, projects, and work products, as well as thousands of interactions. How representative is a supervisor’s perception, and how accurate are her or his critical incidents used as examples? Also, consider how important evaluation is for all of us since it affects our self-image, self-worth, and fundamental security. When you are told you are average or slightly below average, it is hard not to internalize this as being mediocre or inadequate not only as a worker, but as a person too, which understandably triggers self-protective rationalization. Managers can also be prone to introduce distortions. They have time and energy constraints which limit their data collection and the time it takes to diligently and diplomatically navigate the process. Managers generally have a preference to minimize conflict, yet sometimes have pressures to get results no matter what employee needs and constraints may be. Managers also embody their own psychological biases and preferences about who they like and who they do not, which cannot help but filter into the appraisal process. Finally, some systems implicitly discourage robust evaluation, instead opting for wide allowances of all but egregious behavior, while other systems encourage micro-evaluation in a drive for immediate documented results. While the public sector can frequently be an example of the former, in these reform-oriented times it is not atypical to see significant performance standard increases leading to suddenly mandated appraisal systems with robust performance requirements and with the concomitant strengths and weaknesses that go along with such systems. The silly definition of what an appraiser must accomplish, located at the bottom of the slide, does not feel so silly in organizations that do not have an appraisal culture and to individuals who have not been trained in appraisal. This chapter aims to enhance the art and science of appraisal, without ignoring the challenges inherent in the process.

COMMON TYPES OF APPRAISAL (4)

· Trait-based systems

· Behavior-based systems

· Results-based systems

As one commentator noted: “the primary problem supervisors encounter is not knowing who are the best performers, but rather measuring and documenting performance differentials.” Three conceptually different systems can be used; in reality, these systems are often blended together.

5

Trait-based systems focus on the personal characteristics deemed important for the job. Traits tend to be more deeply embedded in the personality structure of individuals, but importantly, individuals can modify them for their work environment. For example, strongly introverted people may need to be conscious of being more friendly in certain business environments. Examples of traits include attitude, friendliness, drive or energy, thoroughness, adaptability, and decisiveness. Traits are widely, and almost instinctively, used. The challenge is that traits tend to have an abstract quality. The trait “attitude” is a classic, because it is at one time characteristic of the importance of traits and the difficulty of evaluating them accurately and fairly. We can all quickly pick out someone who has a “bad attitude” at work and discuss how such poor attitudes dampen morale, impede problem-solving, and pull down productivity. Yet consider the difficulty of providing clear examples that do not seem like personal judgments to recipients of the comments. The trick of all types of appraisal is to provide pertinent, thoughtful and representative examples when there is a deficiency. Such examples must be detailed enough to withstand accusations of vagueness, but generally not so detailed as to seem nick-picking. Of course recitations of egregious behaviors, if and when they should occur, need to be much more detailed and legalistic. At the other end of abstract and broad trait descriptions, one can be more concrete and narrow. This leads to our next category of indicators.

6

A second type of appraisal is behavior-based. Behaviors are generally considered more easily learned behaviors, and less subjective in that they tend to apply to specific aspects of a job. For example, the broad and generic trait, friendliness, might be expressed as customer relations interactions for external contacts and collegiality of internal contacts. Examples of behaviors are written communication, timeliness, customer relations interactions, and accuracy and precision. Behaviors tend to be more concrete and often more job specific. Behaviors are good in terms of job relatedness and specificity as long as they are pertinent and representative. Cataloging systems of behaviors, often called behaviorally anchored rating systems, is widely acknowledged as legally the tightest defensive position, can provide useful specificity, and is easily customized. The challenge to behavioral systems is that they sometimes “miss the forest for the trees” in that the essence of the person’s contributions is lost in the analysis of micro-behaviors which may not be easily defined by subelements. For example, some people have odd ways of getting work done and may even be annoying in completing a project, but the final product is nonetheless exceptional. This leads to the third type of system which focuses on the whole rather than the parts.

7

Results-based systems or outcome-oriented approaches attempt to calibrate one’s contribution to the success of the organization. Such systems often have a management by objectives (MBO) mentality. These systems require mutually agreed-upon goals that are linked to specific, realistic objectives, interim progress reviews, and comparison between actual and expected accomplishments at the end of the rating period. Such systems work best when those being appraised have definable projects or production quotas. Some private sector jobs lend themselves relatively well to this type of system, such as car sales or Wall Street trading, but most line and management functions are a combination of production and human relation skills, where getting to production is through human relations which is notoriously difficult to measure in results-based systems. Also, while goals may emphasize concrete objectives in many cases, such as clients served, or efficiency in completing projects, the basket of goals in the public sector must be sure to capture objectives such as legal and rule-based compliance. Our final cartoon suggests that a total reliance on results based systems may warp the performance of employees.

8

Let’s use the nationally prominent example of teachers to illustrate the challenges of using a pure type of evaluation. Teacher traits such as patience, compassion, and cheerfulness are invariably enormous assets, and being weak in them makes the experience for children far less rich. Behaviors are also very important in the craft of teaching. Examples are lesson plan preparation, the ability to keep concurrent groups learning, the ability to maintain discipline without excessive sternness, engagement of all students including those with learning and behavioral issues, and the ability to spark curiosity and student motivation in learning. Finally, there is the current and contested issue of getting students to make demonstrable progress toward definable goals in nationally standardized tests in key areas such as reading and math. This has been important in trying to jar higher standards, primarily found in low performing schools in economically depressed areas. Many commentators can point out that such areas have much harder challenges in achieving definable results given student’s more limited backgrounds in terms of contributing to education, more troubled family histories, and school’s more limited resources and thus lower salaries. Yet it can also be pointed out that it is inappropriate to abandon such schools to exceptionally low standards as had become a norm for many decades, a norm not found in most other advanced democracies. Setting up a system that includes results-based measures keeps the attention on tackling this difficult problem in which additional financial resources are but a portion of the solution. However, schools must also be community-builders and create cohesive, can-do cultures in areas that are often downtrodden. Thus the debate on how to evaluate teachers is active across the country, as systems try to correctly calibrate the degree to which results on national tests will weigh in their assessments. And nowhere has this debate been more pointed than in troubled districts such as Washington, D.C., Chicago, and Los Angeles.

In reality, most systems use combinations of all three methods. Traits and behaviors are often intermixed using numeric scales and brief comments to cover the areas of time, temperament, team work, flexibility, communications, and compliance/ethics. Production parameters (results) are often also in such lists, but as such, they tend to be generic assessments too. Outcomes oriented assessments are often developed in more detail in separate sections in which description of projects or production goals are discussed and achievements identified. Frequently these discussions are wrapped in improvement or developmental language. Bosses may encourage goal setting in self-assessments, reinforcing some and introducing other goals as necessary in summary evaluative statements.

RATERS AND RATING ERRORS (9)

· Raters: bosses, self, teams, 360 degree assessments, and electronic and automated assessments

· Common rating errors: compatibility error, spill-over effect, recency effect, contrast error, actor-observer bias, leniency error, severity error, central tendency error

· Minimizing errors and distortions

The prime rater in appraisals is normally the boss. Bosses are expected to monitor performance during the year and provide ongoing feedback. This is a major challenge to managers today as their ranks have been slimmed down in recent decades and their responsibilities have proliferated. In the busy work world, many supervisors have a difficult time attending to subordinate problems, much less documenting standard performance. Nonetheless, without diligence in this regard, the appraisal process will suffer substantially. This will matter less if the supervisor is blessed with a relatively consistent high-performing group, but even such individuals like to know that their hard work was appreciated with some specificity. The stimulation of average performers and the prodding of below average performers absolutely demand incisive suggestions born of closer monitoring, analysis, and ongoing feedback.

Yet other raters are possible. Self-evaluative statements are often used prior to formal evaluation. They are very useful at providing self and supervisor insight and for creating a mutual, goal-setting dynamic. In a team based world, the evaluation of team members becomes more and more important. In most public sector settings, this is still streamed through the boss, but can be more emphasized through the use of 360 degree assessments in which feedback is provided by all those who come in contact with the assessee on a regular basis. Like self-assessments, 360 degree assessments are largely developmental but the results can affect the formal process as well. Finally, increasingly there are ways to do electronic monitoring and assessment. Typically such methods only provide some basic production guidelines, such as the old-fashioned time card systems tracked timeliness at work.

Rating errors occur for many reasons stemming from cognitive limitations of raters, intentional manipulation, organizational influences and simple fallibilities of human nature. Common errors include the following:

--The compatibility error is when appraisal is biased because someone is like us and compatibility influences the rater.

--The spillover effect is bias introduced because of an exceptional event. Because of a successful project, important deficiencies are overlooked, or because of a major lapse, the employee is judged more harshly in other areas as well.

--The recency effect is powerful because we remember the latest events much better than the past, and psychologically, unless diligent record keeping has occurred, the effect of a recent problem or success will be given significantly more weight.

--The contrast effect becomes an error when the peer group norms are used rather than objective criteria. This can be a problem, for example, in a high-performing group when an “average” member is judged harshly, perhaps because of their heavy family responsibilities.

--The actor-observer bias is when external factors are used as scapegoats for poor performance, even though these factors may be chronic and/or anticipated.

--The leniency error exists when supervisors want to avoid hurting people’s feelings or simply want to avoid unpleasantness or possible confrontations by aggressive employees. This is quite common in unionized and governmental contexts in which job property and civil service rights strengthen employees’ rights and embolden outspoken employees.

--The severity error can occur when a unit has experienced a negative evaluation, the supervisor is obsessive in his or her standards expectations, or there is an interest in shocking an average employee into higher performance.

--The error of central tendency is to clump most employees in the middle. This is also common for a variety of reasons. For example, when appraisal systems require documentation of very high or very low evaluations, it encourages the central tendency effect. This is not necessarily a bad thing to the degree that the bell curve is realistically operating in the workplace as it often does. However, the error of central tendency can occur more frequently because of managers’ inability to maintain good records of performance, and thus it is easier to push everyone to the middle of the rating scales.

10

Improving the Process by Reducing Errors and Increasing Structured Interaction

Minimizing errors occurs at different levels. Bosses can minimize errors by making evaluation a year-long process and by having information data collection systems that cover all employees. They can engage employees in proactive discussions before the annual evaluation process, so that there are no surprises. In our cartoon, we can see that the boss was late in gathering data, and then tried to cover it up by being participative, …but the employees saw through the ruse. Bosses can take seriously the idea that half of an employee’s failure belongs to the supervisor who inadequately placed, directed, supported, or corrected employees in the HR process. Self-awareness by bosses can mitigate against common human biases such as the compatibility error. Our next cartoon illustrates that evaluation is tough for bosses too. Organizations can be aware of the supervisor incentives that they build into evaluation systems, and adjust such systems to take into account internal distortions and external realities. Training on evaluation and assessment can make an enormous difference in the attention to appraisal as well as its quality. Finally, employees themselves can be proactive in seeking feedback from bosses and others and making an effort to improve areas deemed weaker or inadequate, or simply by making sure that good performance is well documented and shared with superiors.

11

The Ideal Process

To summarize, the ideal evaluation can constitute 7 elements:

· Goal setting

· Alignment of individual and organizational goals

· Self-assessment

· 360 reviews

· managerial appraisal

· competency assessment and

· developmental planning

DISCIPLINARY SYSTEMS (12)

· Self-discipline

· Disciplinary factors to consider

· Progressive discipline or positive discipline

Most employees know the rules and use self-discipline to conform to them. When provided with legitimate feedback and evidence of suboptimal performance or minor breaches in protocol, they willingly exercise corrective action on their own. Supervisors want to harness self-discipline to the greatest degree for both the individual and organizational benefit.

When poor performance is more substantial and/or sustained, or when misconduct is occurring, discipline is needed. Because of employee property rights related to due process in adverse actions, supervisors must be careful to follow rules scrupulously, not jump to conclusions without evidence and inquiry, and document well. For example, it is easy to hear a report of an infraction that sounds egregious, and to take immediate action, only to find that there were extraordinary circumstances, or worse, manipulation of innocent data to make someone else look badly. Such an unfortunate incident occurred to President Obama—normally a man known to be careful in making personnel judgments—in the hasty firing of a Department of Agriculture employee who was supposedly “proved” to be using racial stereotyping on videotape. Only later was the context of the comments provided, and was it subsequently learned that the video had been carefully edited to heighten misrepresentation of the speaker’s actual point. The President had to personally apologize to the employee because he had been a party to the “rush-to-judgment.”

Factors that are normally considered in the disciplinary process include problem severity, duration, extenuating circumstances, organizational policies and employee training, agency past practice, and the degree of management support one is likely to receive.

Progressive discipline is a method to ensure that poor performance or minor misconduct is treated in a rational fashion so that workers know what the problem is, what change is expected, and the consequences of inaction. Problem identification starts with informal counseling. Continued problems lead to a verbal warning. When the severity or duration indicates the need, the discipline is elevated to a written reprimand that may be for documentation or the official file. Continued problems may lead to suspensions, and potentially at some point, to separation. Positive discipline emphasizes the need for supervisors to be proactive in a problem solving mode and as supportive as possible, without sacrificing organizational interests.

It should be noted that major misconduct may require an adverse action without preliminary steps found in progressive discipline, and in extreme cases, the rules may allow and the situation may necessitate moving immediately to separation.

13

CONCLUSION

In conclusion appraisal is considered to be unpleasant to most supervisors, subject to distortions and error, and poorly viewed by most employees. Even very good employees who get excellent evaluations may be upset because others with far inferior performance get nearly identical ratings because of evaluation inflation or leniency effects. Nonetheless, the underlying functions of development, evaluation, standards enforcement and systems evaluation depend on the information generated in the appraisal process. No matter how imperfect evaluation systems may be, the converse, not doing any appraisal, is unacceptable. Indeed, despite the challenges, many supervisors use appraisal very effectively as a long-term tool to proactively improve the culture and productivity of the organization.